检索结果-内蒙古大学图书馆

Coupling streaming AI and HPC ensembles to achieve 100-1000× faster biomolecular simulations

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Brace, Alexander Yakushin, Igor Ma, Heng Trifan, Anda Munson, Todd Foster, Ian Ramanathan, Arvind Lee, Hyungro Turilli, Matteo Jha, Shantenu Data Science and Learning Division Mathematics and Computer Science Division Argonne National Laboratory Lemont United States Department of Computer Science University of Chicago Chicago United States Center For Biophysics University of Illinois Urbana-Champaign Champaign United States RADICAL-Lab Rutgers University New Brunswick United States Computational Science Initiative Brookhaven National Laboratory Upton United States

Machine learning (ML)-based steering can improve the performance of ensemble-based simulations by allowing for online selection of more scientifically meaningful computations. We present DeepDriveMD, a framework for ML-driven steering of scientific simulations that we have used to achieve orders-of-magnitude improvements in molecular dynamics (MD) performance via effective coupling of ML and HPC on large parallel computers. We discuss the design of DeepDriveMD and characterize its performance. We demonstrate that DeepDriveMD can achieve between 100–1000× acceleration for protein folding simulations relative to other methods, as measured by the amount of simulated time performed, while covering the same conformational landscape as quantified by the states sampled during a simulation. Experiments are performed on leadership-class platforms on up to 1020 nodes. The results establish DeepDriveMD as a high-performance framework for ML-driven HPC simulation scenarios, that supports diverse MD simulation and ML back-ends, and which enables new scientific insights by improving the length and time scales accessible with current computing capacity. © 2021, CC BY.

关键词： Molecular dynamics

Learner Action Patterns in the Problem-Solving Process Related to Program Code Composition Based on Tracking System Activities

学校读者我要写书评

暂无评论

Learner Action Patterns in the Problem-Solving Process Relat...

International Conference on Informatics and Computing (ICIC)

作者： Aulia Akhrian Syahidi Ahmad Afif Supianto Tsukasa Hirashima Yutaka Watanobe Smart City Information System Applied Undergraduate Program Politeknik Negeri Banjarmasin Banjarmasin City Indonesia Research Center for Data and Information Sciences National Research and Innovation Agency (BRIN) Bandung City Indonesia Learning Engineering Laboratory Graduate School of Advanced Science and Engineering Hiroshima University Kagamiyama Higashihiroshima Japan Department of Computer Science and Engineering Division of Information Systems University of Aizu Aizu-Wakamatsu City Fukushima Japan

ISBN: (纸本)9798350345728

learning activities are an indicator of the learner's desire to learn during the learning process. The pattern of learner action is related to learning activities. In this case, in extracting the learning process, it is necessary to collect a lot of data through analysis of the learning process. The purpose of this study is to recommend and report the performance of an activity tracking system equipped with visual artifacts as an educational data mining approach to analyze action patterns when learners complete the arrangement of program code lines in each programming problem. In this study, data were taken from activity recordings when learners used interactive learning media for a basic programming subject called TOLSYASUPI-EduMed which had problem-solving learning models embedded in their interactions. learning Analytics is used as a method of this research and data relationships from learner actions during the learning process using the “if-then” rule. Log data is used for recording and detecting activities carried out by learners. We confirm that the results of the study are based on the findings of the data relationship that if the lines of program code are too long (more than five lines) then almost all learners experience a bottleneck condition. The completion time of each question and the number of iterations in completing the program code lines for each question have also been recorded and have a pattern of connection between the data. The information that has been obtained is then forwarded to teachers or stakeholders as key information to make appropriate feedback to learners based on the results as a means of analyzing and evaluating the programming learning process.

关键词： Visualization Codes Media data models Recording data mining Problem-solving

A fast accurate fine-grain object detection model based on YOLOv4 deep neural network

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Roy, Arunabha M. Bose, Rikhi Bhaduri, Jayabrata University of Michigan Aerospace Engineering Ann ArborMI48109 United States Mechanical Engineering Johns Hopkins University BaltimoreMD21218 United States Capacloud AI Deep Learning & Data Science Division WB Kolkata711103 India

Early identification and prevention of various plant diseases in commercial farms and orchards is a key feature of precision agriculture technology. This paper presents a high-performance real-time fine -grain object detection framework that addresses several obstacles in plant disease detection that hinder the performance of traditional methods, such as, dense distribution, irregular morphology, multi-scale object classes, textural similarity, etc. The proposed model is built on an improved version of the You Only Look Once (YOLOv4) algorithm. The modified network architecture maximizes both detection accuracy and speed by including the DenseNet in the back-bone to optimize feature transfer and reuse, two new residual blocks in backbone and neck enhance feature extraction and reduce computing cost;the Spatial Pyramid Pooling (SPP) enhances receptive field, and a modified Path Aggregation Network (PANet) preserves fine-grain localized information and improve feature fusion. Additionally, use of the Hard-Swish function as the primary activation improved the model’s accuracy due to better nonlinear feature extraction. The proposed model is tested in detecting four different diseases in tomato plants under various challenging environments. The model outperforms the existing state-of-the-art detection models in detection accuracy and speed. At a detection rate of 70.19 FPS, the proposed model obtained a precision value of 90.33%, F1-score of 93.64%, and a mean average precision (mAP) value of 96.29%. Current work provides an effective and efficient method for detecting different plant diseases in complex scenarios that can be extended to different fruit and crop detection, generic disease detection, and various automated agricultural detection *** Codes 68T01, 68T05, 68T07, 68T10, 68T40, 68T45, 68U10, Copyright © 2021, The Authors. All rights reserved.

关键词： Object recognition

GauSS-MI: Gaussian Splatting Shannon Mutual Information for Active 3D Reconstruction

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Xie, Yuhan Cai, Yixi Zhang, Yinqiang Yang, Lei Pan, Jia School of Computing and Data Science The University of Hong Kong Hong Kong Division of Robotics Perception and Learning KTH Royal Institute of Technology Stockholm Sweden Faculty of Engineering The University of Hong Kong Hong Kong Centre for Transformative Garment Production Hong Kong

This research tackles the challenge of real-time active view selection and uncertainty quantification on visual quality for active 3D reconstruction. Visual quality is a critical aspect of 3D reconstruction. Recent advancements such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have notably enhanced the image rendering quality of reconstruction models. Nonetheless, the efficient and effective acquisition of input images for reconstruction—specifically, the selection of the most informative viewpoint—remains an open challenge, which is crucial for active reconstruction. Existing studies have primarily focused on evaluating geometric completeness and exploring unobserved or unknown regions, without direct evaluation of the visual uncertainty within the reconstruction model. To address this gap, this paper introduces a probabilistic model that quantifies visual uncertainty for each Gaussian. Leveraging Shannon Mutual Information, we formulate a criterion, Gaussian Splatting Shannon Mutual Information (GauSS-MI), for real-time assessment of visual mutual information from novel viewpoints, facilitating the selection of next best view. GauSS-MI is implemented within an active reconstruction system integrated with a view and motion planner. Extensive experiments across various simulated and real-world scenes showcase the superior visual quality and reconstruction efficiency performance of the proposed system. Copyright © 2025, The Authors. All rights reserved.

关键词： Gaussian distribution

BraggNN: Fast X-ray Bragg peak analysis using deep learning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Liu, Zhengchun Kenesei, Peter Sharma, Hemant Miceli, Antonino Park, Jun-Sang Almer, Jonathan Kettimuthu, Rajkumar Foster, Ian Data Science and Learning Division Argonne National Laboratory LemontIL60564 United States X-Ray Science Division Argonne National Laboratory LemontIL60439 United States

X-ray diffraction based microscopy techniques such as High Energy Diffraction Microscopy rely on knowledge of the position of diffraction peaks with high precision. These positions are typically computed by fitting the observed intensities in area detector data to a theoretical peak shape such as pseudo-Voigt. As experiments become more complex and detector technologies evolve, the computational cost of such peak detection and shape fitting becomes the biggest hurdle to the rapid analysis required for real-time feedback during in-situ experiments. To this end, we propose BraggNN, a deep learning-based method that can determine peak positions much more rapidly than conventional pseudo-Voigt peak fitting. When applied to a test dataset, BraggNN gives errors of less than 0.29 and 0.57 pixels, relative to the conventional method, for 75% and 95% of the peaks, respectively. When applied to a real experimental dataset, a 3D reconstruction that used peak positions computed by BraggNN yields 15% better results on average as compared to a reconstruction obtained using peak positions determined using conventional 2D pseudo-Voigt fitting. Recent advances in deep learning method implementations and special-purpose model inference accelerators allow BraggNN to deliver enormous performance improvements relative to the conventional method, running, for example, more than 200 times faster than a conventional method on a consumer-class GPU card with out-of-the-box software. Copyright © 2020, The Authors. All rights reserved.

关键词： X ray diffraction

Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kirchhoff, Yannick Rokuss, Maximilian R. Roy, Saikat Kovacs, Balint Ulrich, Constantin Wald, Tassilo Zenk, Maximilian Vollmuth, Philipp Kleesiek, Jens Isensee, Fabian Maier-Hein, Klaus Heidelberg Division of Medical Image Computing Germany Faculty of Mathematics and Computer Science Heidelberg University Germany HIDSS4Health - Helmholtz Information and Data Science School for Health Heidelberg Karlsruhe Germany Medical Faculty Heidelberg Heidelberg University Heidelberg Germany Helmholtz Imaging German Cancer Research Center Heidelberg Germany Clinic for Neuroradiology University Hospital Bonn Bonn Germany Medical Faculty Bonn University of Bonn Bonn Germany University Hospital Essen Essen Germany West German Cancer Center Essen University Hospital Essen Essen Germany Pattern Analysis and Learning Group Department of Radiation Oncology Heidelberg University Hospital Germany

Accurately segmenting thin tubular structures, such as vessels, nerves, roads or concrete cracks, is a crucial task in computer vision. Standard deep learning-based segmentation loss functions, such as Dice or Cross-Entropy, focus on volumetric overlap, often at the expense of preserving structural connectivity or topology. This can lead to segmentation errors that adversely affect downstream tasks, including flow calculation, navigation, and structural inspection. Although current topology-focused losses mark an improvement, they introduce significant computational and memory overheads. This is particularly relevant for 3D data, rendering these losses infeasible for larger volumes as well as increasingly important multi-class segmentation problems. To mitigate this, we propose a novel Skeleton Recall Loss, which effectively addresses these challenges by circumventing intensive GPU-based calculations with inexpensive CPU operations. It demonstrates overall superior performance to current state-of-the-art approaches on five public datasets for topology-preserving segmentation, while substantially reducing computational overheads by more than 90%. In doing so, we introduce the first multi-class capable loss function for thin structure segmentation, excelling in both efficiency and efficacy for topology-preservation. Our code is available to the community, providing a foundation for further advancements, at: https://***/MIC-DKFZ/Skeleton-Recall. © 2024, CC BY.

关键词： Topology

A generative artificial intelligence framework based on a molecular diffusion model for the design of metal–organic frameworks for carbon capture

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Park, Hyun Yan, Xiaoli Zhu, Ruijie Huerta, Eliu A. Chaudhuri, Santanu Cooper, Donny Foster, Ian Tajkhorshid, Emad Data Science and Learning Division Argonne National Laboratory LemontIL60439 United States Theoretical and Computational Biophysics Group NIH Resource Center for Macromolecular Modeling and Visualization Beckman Institute for Advanced Science and Technology University of Illinois at Urbana-Champaign UrbanaIL61801 United States Center for Biophysics and Quantitative Biology University of Illinois at Urbana-Champaign UrbanaIL61801 United States Multiscale Materials and Manufacturing Lab University of Illinois Chicago ChicagoIL60607 United States Department of Materials Science and Engineering Northwestern University EvanstonIL60208 United States Department of Computer Science University of Chicago ChicagoIL60637 United States Department of Physics University of Illinois at Urbana-Champaign UrbanaIL61801 United States Computational Science and Engineering Data Science and AI Department TotalEnergies EP Research & Technology USA LLC HoustonTX77002 United States Department of Biochemistry University of Illinois at Urbana-Champaign UrbanaIL61801 United States

Metal–organic frameworks (MOFs) exhibit great promise for CO2 capture. However, finding the best performing materials poses computational and experimental grand challenges in view of the vast chemical space of potential building blocks. Here, we introduce GHP-MOFassemble, a generative artificial intelligence (AI), high performance framework for the rational and accelerated design of MOFs with high CO2 adsorption capacity and synthesizable linkers. GHP-MOFassemble generates novel linkers, assembled with one of three pre-selected metal nodes (Cu paddlewheel, Zn paddlewheel, Zn tetramer) into MOFs in a primitive cubic topology. GHP-MOFassemble screens and validates AI-generated MOFs for uniqueness, synthesizability, structural validity, uses molecular dynamics simulations to study their stability and chemical consistency, and crystal graph neural networks and Grand Canonical Monte Carlo simulations to quantify their CO2 adsorption capacities. We present the top six AI-generated MOFs with CO2 capacities greater than 2m molg−1, i.e., higher than 96.9% of structures in the hypothetical MOF dataset. Copyright © 2023, The Authors. All rights reserved.

关键词： Carbon dioxide

Deep learning-based low-dose Tomography reconstruction with hybrid-dose measurements

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Wu, Ziling Bicer, Tekin Liu, Zhengchun de Andrade, Vincent Zhu, Yunhui Foster, Ian T. Bradley Department of Electrical & Computer Engineering Virginia Polytechnic Institute State University United States Data Science & Learning Division X-Ray Science Division Argonne National Laboratory United States

Synchrotron-based X-ray computed tomography is widely used for investigating inner structures of specimens at high spatial resolutions. However, potential beam damage to samples often limits the X-ray exposure during tomography experiments. Proposed strategies for eliminating beam damage also decrease reconstruction quality. Here we present a deep learning-based method to enhance low-dose tomography reconstruction via a hybrid-dose acquisition strategy composed of extremely sparse-view normal-dose projections and full-view low-dose projections. Corresponding image pairs are extracted from low-/normal-dose projections to train a deep convolutional neural network, which is then applied to enhance full-view noisy low-dose projections. Evaluation on two experimental datasets under different hybrid-dose acquisition conditions show significantly improved structural details and reduced noise levels compared to uniformly distributed acquisitions with the same number of total dosage. The resulting reconstructions also preserve more structural information than reconstructions processed with traditional analytical and regularization-based iterative reconstruction methods from uniform acquisitions. Our performance comparisons show that our implementation, HDrec, can perform denoising of a real-world experimental data 410x faster than the state-of-the-art Xlearn method while providing better quality. This framework can be applied to other tomographic or scanning based X-ray imaging techniques for enhanced analysis of dose-sensitive samples and has great potential for studying fast dynamic processes. Copyright © 2020, The Authors. All rights reserved.

关键词： Image reconstruction

Applications of physics informed neural operators

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Rosofsky, Shawn G. Majed, Hani Al Huerta, E.A. Data Science and Learning Division Argonne National Laboratory LemontIL60439 United States Department of Physics University of Illinois at Urbana-Champaign UrbanaIL61801 United States NCSA University of Illinois at Urbana-Champaign UrbanaIL61801 United States Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign UrbanaIL61801 United States Department of Computer Science The University of Chicago ChicagoIL60637 United States

We present an end-to-end framework to learn partial differential equations that brings together initial data production, selection of boundary conditions, and the use of physics-informed neural operators to solve partial differential equations that are ubiquitous in the study and modeling of physics phenomena. We first demonstrate that our methods reproduce the accuracy and performance of other neural operators published elsewhere in the literature to learn the 1D wave equation and the 1D Burgers equation. Thereafter, we apply our physics-informed neural operators to learn new types of equations, including the 2D Burgers equation in the scalar, inviscid and vector types. Finally, we show that our approach is also applicable to learn the physics of the 2D linear and nonlinear shallow water equations, which involve three coupled partial differential equations. We release our artificial intelligence surrogates and scientific software to produce initial data and boundary conditions to study a broad range of physically motivated scenarios. We provide the source code, an interactive website to visualize the predictions of our physics informed neural operators, and a tutorial for their use at the data and learning Hub for science. © 2022, CC BY.

关键词： Boundary conditions